CN1356778A - Encode/decode method and system of modulation-array code for bulk holographic memory - Google Patents

Abstract

An encode-decode method for the modulation-array code used by bulk holographic memory features that its encode method is the combination of modulation encoded method and error-correcting encode method and its decode method based on the soft-judge metric algorithm uses an iterative algorithm. Its encode system is composed of the encoder of 6:8 equivalent weight code and the encoder of systemetic error-correcting code. Its decode system is composed of memory crystal, CCD, filter and computer.

Description

Volume hologram memory volume, interpretation method and the system of modulation-array code

The present invention relates to the body holographic storage technology, designed a kind of volume, interpretation method and system of modulation-array code.

The theoretical limit of body Hologram Storage is V/ λ ³(V is the recording medium volume, and λ is an operation wavelength), its theoretical storage density can reach 10 ¹²Bits/cm ³The order of magnitude; And owing to light beam carries pictorial information, write to read to be the parallel processing of 2-D data page or leaf, can reach high storage and reading speed.Therefore body hologram memory system becomes the important development direction of area information storage with its salient feature and potentiality, and has obtained certain achievement in the last few years.Yet complicated noise jamming is the significant obstacle in its practicability process always in the system.

Not only the kind of noise is a lot of in the body hologram memory system, and to different storage mediums, system configuration, multiplex technique and data access package, and various noises produce influences difference, and the noise source that finally has the decision influence is also inequality.These characteristics of noise have determined to carry out simple threshold decision can not satisfy the requirement of user to the error rate.Therefore, must introduce error correction coding and modulating-coding technology, but the coding structure of modulation code commonly used has determined the polarity that only can utilize a part of input information to finish code element to judge when decoding.Therefore in decode procedure, can lose a large amount of information, be unfavorable for improving the information utilization of system.Being reflected to the storage aspect is exactly the raising that is unfavorable for effective storage density.And traditional error correcting code and modulation code in use all be independently, thereby the modulated code decoder of bulk information of data carry is blocked, and the error correcting code decoder can not make full use of the entrained information of data that receives.

Usually the soft-decision metrics that code element u is traditional is represented than L (u) through log-likelihood function commonly used: $L\left(u\right)=\log \frac{p(u=0)}{p(u=1)}$

The conditional log likelihood function of code element u ratio is under the condition that obtains measured value y:

L(u|y)＝L(y|u)+L(u)＝Lc×y+L(u)? $L\left(y\right|u)=\log \frac{p\left(y\right|u=0)}{p\left(y\right|u=1)}$

For separate code element u1 and u2 following operation rule is arranged:

L(u1u2)≈sgn(L(u1))×sgn(L(u2))×min(|L(u1)|，|L(u2)|)

Wherein P represents probability, and sgn () is-symbol function, min () are to get minimum value function, and Lc represents the various complex informations of channel.Because noise parameter changes with the space in the body hologram memory system, thus Lc also change with the space, thereby make L (u|y) be difficult to calculate, therefore the signal to noise ratio at channel has fluctuating, and result of use is not good in the noise parameter communication channel that is difficult to add up.

The objective of the invention is to overcome aforementioned disadvantages, a kind of new coding method, interpretation method and system thereof that is applicable to that volume hologram memory reads and writes data and handles is provided, thereby can make full use of the amount of information that signal carries, finally reduce user's error rate, improve real system memory capacity.

The design of overall plan of the present invention is, proposes a kind of coding structure that is applicable to that volume hologram memory reads and writes data and handles, and it is the coding techniques that a kind of modulation code and error correcting code are combined closely, and it is different from traditional cascaded code that is formed by two kinds of error correcting code cascades.Our weighing-appliance has the sign indicating number of this coding structure to be modulation-array code.Used the soft-decision metrics method that adapts with it simultaneously, comprehensive modulation code detection on this basis, soft-decision, array code error correction and Turbo code decoding principle have formed a kind of new iterative decoding algorithm.Concrete technical scheme of the present invention is expressed as follows:

The coding method of modulation-array code that a kind of volume hologram memory of the present invention is used contains modulation coding method and error correction/encoding method, it is characterized in that it is made of in turn following steps:

1), make the storage data carry out the equivalent weight code coding by the equivalent weight code encoder;

2), carry out error correcting code coding as input by the error correcting code encoder with equivalent weight code, the error correcting code of employing is a systematic code;

3), code element structure behind the coding from left to right is made up of modulation code part and verification unit part successively, and wherein, left side modulation code partly is: k * k code element from left to right is: { x (1,1), x (1,2) ... x (1, k) } ... { x (k, 1), x (k, 2) ... x (k, k) } be that k n:k modulation code code word arranged in order, part: the k of the right verification unit * 2 code elements, wherein, h (1), h (2) ... h (k-1), the pass of h (k) and left side sign indicating number is: h (i)=x (1, i) x (2, i) ... x (k-1, i) x (k, i) 1 i=1,2 ... k-1, k, d (1), d (2) ... d (k-1), d (k), with the pass of left side sign indicating number be: d (i)=f (1, i) f (2, i) ... f (k-1, i) f (k, i) 1 i=1,2 ... k-1, k, f (j, i)=x (j, ((i+j-2) mod k)+1) j=1,2 ... k-1, k.

Referring to the installation drawing of Fig. 3 code device, the coding method of modulation-array code of using according to aforesaid volume hologram memory is characterized in that: equivalent weight code adopts 6: 8 equivalent weight codes, and this encoder is more common, is commercially available.Systematic error checking code adopts the modified model array code of being made up of two (9,8) parity check codes, and this encoder is because coding structure changes, and the coding structure design of pressing Fig. 1 C with dsp chip realizes that DSP is general.

The coded system that the coding method of modulation-array code of using according to aforesaid volume hologram memory designs, it is characterized in that: it contains oriented 6: 8 equivalent weight code encoders provides the data source of storing data, 6: 8 equivalent weight code encoders of coded data are provided to the systematic error checking code encoder, provide the systematic error checking code encoder of coded data, the spatial light modulator of the coded data after storage crystal provides modulation and storage crystal to spatial light modulator.

The interpretation method of modulation-array code that a kind of volume hologram memory of the present invention is used, contain the soft-decision metrics algorithm, it is characterized in that: this decode procedure has used the soft-decision metrics method that adapts with its coding structure, comprehensive modulation code detection on this basis, soft-decision, array code error correction and Turbo code decoding principle, formed a kind of new iterative decoding technology, it is made up of following steps successively:

1), the data of coming from storage crystal are read in data card;

2), computer is read a code word from data card;

3), enter the decoding subprogram;

4), be the code element of the modulation code in code word part that unit sorts from big to small by the amplitude that receives with the equivalent weight code code word, add up overall discrimination threshold;

5), enter the soft-decision metrics counting subroutine: at first the separated into two parts code element is calculated respectively, wherein modulation code part code element enters modulation code part soft-decision metrics primary Calculation subprogram, and verification unit part code element enters verification unit part soft-decision metrics counting subroutine.The two enters modulation code part code element soft-decision metrics patch routine after calculating and finishing;

6), 5) in the modulation code part soft-decision metrics primary Calculation subprogram mentioned, wherein the primary Calculation formula is:

S(u|y)＝c×|b+d×y|

C representative is 0 or 1 o'clock positive-negative polarity information with bit decision in the formula, and d and b represent the various complex informations of channel,

If: receiving data maximum after putting in order is Max, and minimum value is 0, and then: boundary condition is:

When judging u=1, if the confidence level minimum that y=0 then judges, | S (u|y) |=0,

If the confidence level maximum that y=Max then judges, | S (u|y) |=Max, when judging u=0, if the confidence level maximum that y=0 then judges, | S (u|y) |=Max,

If the confidence level minimum that y=Max then judges, | S (u|y) |=0, draw:

C in the formula ₁, c ₀The expression code element get 1 and 0 o'clock S (u|y) polarity, c wherein ₁* c ₀=-1;

7) the verification unit part soft-decision metrics counting subroutine of, 5) being mentioned, wherein threshold value is 4) in the global decision threshold value mentioned, in modulation code part code element sequencer procedure, take out the m1 and (m1-1) range value of individual code element in each equivalent weight code code word, search out the threshold value that makes these code elements differentiate back error code minimum, the best global threshold of the modulation code part code element that Here it is counts, it is approximated to the decision threshold of verification unit part code element, so verification unit part soft-decision metrics computing formula is:

8), 5) in the modulation code part code element soft-decision metrics patch routine mentioned divided for three steps carried out at least, wherein:

At first enter the soft-decision metrics first step patch routine of modulation code part code element, revise according to the position after the ordering in the modulation code word successively:

Wherein: f (y, n)=y * g (n) n is the position after code element sorts in modulation code word, n=0, and 1,2 ..., k-1,

Then enter second step of modulation code part code element soft-decision metrics patch routine, revised according to the bit error probability of code element:

S _m1-1(u|y)＝c ₁×min(|S _m1-1(u|y)|，|S _m1(u|y)|)

S _m1(u|y)＝c ₀×|S _m1-1(u|y)|

M1 represents in the modulation code code word 1 number, S in the formula _n(u|y) represent the soft-decision metrics of n code element, n is the position after code element sorts in modulation code word, n=0, and 1,2 ..., k-1,

Carry out the 3rd step of the soft-decision metrics patch routine of modulation code part code element then, revise according to the reception range value of code element: as n, n+1 ..., the y value of n+m code element is identical, and this m+1 code element polarity is identical, then:

S _n(u|y)=...=S _N+m(u|y)=sgn (S _n(u|y)) * min (| S _n(u|y) | ..., | S _N+m(u|y) |) if: the polarity difference of m+1 code element, then:

S _n(u|y)=S _N+1(u|y)=...=S _N+m(u|y) thereby=0 obtain the condition soft-decision metrics value of modulation code part code element;

9), carry out the iterative decoding of code word, wherein the operation rule of soft-decision metrics is as follows:

S (u1 u2) ≈ sgn (S (u1|y1)) * sgn (S (u2|y2)) * min (| S (u1|y1) |, | S (u2|y2) |) sgn () gets sign function in the formula, min () is the function of minimizing;

10), quantize according to the final soft-decision metrics of code element, its polarity is c ₁Code element be 1, polarity is c ₀Code element be 0, and then it is carried out equivalent weight code decoding, the court verdict data after the output decoding at last;

11), judgement finishes to obtain storing data after data card will be adjudicated output.

Wherein 5) the soft-decision metrics counting subroutine in can be found out by the program circuit of Fig. 8 and Fig. 9: the ownership of the soft-decision metrics basis code element of code element is calculated respectively, the code element of modulation code part enters modulation code part soft-decision metrics primary Calculation subprogram, the code element of verification unit part enters verification unit part soft-decision metrics counting subroutine, after the soft-decision metrics of all code elements calculates end, carry out modulation code part code element soft-decision metrics patch routine again.Revise and divided for three steps carried out at least, idiographic flow is referring to Figure 12～Figure 15.We have just obtained the soft-decision metrics of all code elements to have revised the back, by Fig. 8 and Fig. 9 as can be known program turn back to the iterative decoding subprogram and carry out iterative decoding.

The decoding system that the interpretation method of modulation an array sign indicating number of using according to aforesaid volume hologram memory designs, it is characterized in that, it comprises storage crystal, scan the charge coupled device ccd that receives data from storage crystal, filter filtering is carried out in output to CCD, the computer that links to each other with filter output successively, computer have and receive and the data card of dateout.

Kept the advantage that the conventional modulated sign indicating number is easy to judge receiving symbol polarity by aforementioned visible the present invention, introduced error correction coding simultaneously, made it have error correcting capability.Can adopt iterative decoding algorithm during decoding, improve error correcting capability greatly.When adopting Soft decision decoding, can make full use of the information that data comprise,, adjust the soft value of condition of code element, output error rate is further reduced promptly by information from the modulation code acquisition.

The positive and negative value of soft-decision metrics of the present invention is by making full use of codeword structure of the present invention: modulation code part code element is made up of the equivalent weight code code word, thereby utilizes some boundary conditions can obtain the soft-decision metrics of code element by simple calculating.As seen in computational process, walked around the concrete calculating of channel statistical information, but the result can embody channel statistic property again, so be equivalent to can be dynamically according to the actual channel statistical parameter and adaptive corrected Calculation parameter, so soft-decision metrics of the present invention can have fluctuating in the signal to noise ratio of channel, and use in the channel that noise parameter is difficult to add up, and the polarity and the confidence level of more accurate reflection code element.Therefore the new soft-decision metrics among the present invention not only is applicable to body hologram memory system, and it can also have fluctuating in the signal to noise ratio of channel, and can use in the communication channel that noise parameter is difficult to add up.

Description of drawings:

Figure 1A: according to the theory diagram of coding method proposed by the invention, 1, the storage data, 2, the equivalent weight code encoder, 3, the systematic error checking code encoder, 4, coded data;

Figure 1B: the theory diagram of traditional cascaded code, 5, the first error correcting code encoder, 6, the second error correcting code encoder;

Fig. 1 C: according to symbols encoded structure chart proposed by the invention, 7, the modulation code part, 8, verification unit part;

Fig. 2: body hologram memory system block diagram, 9, data source, 10, spatial light modulator, 11, storage crystal, 12, CCD, 13, filter, 14, dateout, 15, computer, 24, data card, A, coding, B, decoding;

Fig. 3: the installation drawing of code device of the present invention, 16,6: 8 equivalent weight code encoders, 17, buffer memory, 18, (9,8) parity check code coder, 19, modified model array code encoder;

Fig. 4: cataloged procedure of the present invention simplify coding schematic diagram, C, initial data, D, modulation code, E, modulation-array code;

Fig. 5: code translator functional-block diagram, 20, receive data, 21, the soft-decision metrics calculator, 22, the soft-decision metrics corrector, 23, iterative decoding and bit decision device;

Fig. 6: the installation drawing of realizing decode procedure in the body hologram memory system;

Fig. 7: the main program flow chart of decoding;

Fig. 8: decoding subroutine flow chart;

Fig. 9: soft-decision metrics counting subroutine flow chart;

Figure 10: modulation code part soft-decision metrics primary Calculation subroutine flow chart;

Figure 11: verification unit part soft-decision metrics counting subroutine flow chart;

Figure 12: modulation code part code element soft-decision metrics first step patch routine flow chart;

Figure 13: second step of modulation code part code element soft-decision metrics patch routine flow chart;

Figure 14: the 3rd step of modulation code part code element soft-decision metrics patch routine flow chart;

Figure 15: press start, end patch routine flow chart;

Figure 16: preliminary experimental result picture, the original error rate after overall optimal threshold quantizes is adopted in G, input, output error rate after F, the decoding, the result of R1, employing 8 * 10 modulation-array codes, the result of R2,1: 2 modulation code, the result of R3,6: 8 modulation codes, adopts overall optimal threshold gained result at R4.

Table 1: initial data;

Table 2:6: the data behind the 8 equivalent weight codes coding;

Table 3: table 2 data are carried out data after the error correction coding, just the data behind table 1 data-modulated-array code coding;

Table 4: receive data;

Table 5: code element ordering back data (the numeral code element at code element end descending order in code word, example-0 expression is maximum, and-1 expression is second largest ,-7 expressions are minimum);

Table 6: soft-decision metrics primary Calculation result (can draw threshold value by table 5 is 134);

Table 7: the revised result of the modulation code part code element soft-decision metrics first step;

Table 8: revised result of second step of modulation code part code element soft-decision metrics;

Table 9: revised result of the 3rd step of modulation code part code element soft-decision metrics;

Table 10: result behind the iterative decoding;

Table 11: bit decision output result.

Now in conjunction with the embodiments method and apparatus proposed by the invention is done detailed describing.

Adopt aforesaid technical scheme design of the present invention.Wherein the installation drawing of code device is asked for an interview Fig. 3, and equivalent weight code adopts 6: 8 equivalent weight codes, and this encoder is commercially available.Systematic error checking code adopts the modified model array code of being made up of two (9,8) parity check codes, and the coding structure design that this encoder adopts dsp chip to press Fig. 1 C realizes that DSP also is commercially available.

For the ease of clear description cataloged procedure, simplify coding as shown in Figure 4, wherein equivalent weight code is selected 2: 4 equivalent weight codes.Parity check code is (5,4) sign indicating number.

The code translator functional-block diagram is asked for an interview Fig. 5.Fig. 6 is an installation drawing of realizing decode procedure in body hologram memory system.Wherein data card is mainly finished the input and output of data to computer, does not have too high request, is commercially available.What computer was used is the microcomputer of PIII800.

Whole implementation processes of this interpretation method are seen Fig. 6～Figure 15.

The main process of decoding is realized that by computer software the main program flow chart of its realization is shown in 6.The symbol amplitude value that wherein measures is y, and its maximum is Max, and minimum value is 0.

Describe concrete volume, the decode procedure of one group of data in detail below in conjunction with table 1-table 11.

Cataloged procedure: table 1 is an initial data, obtains table 2 result through 6: 8 behind the equivalent weight code coding.For example: first columns of table 1 according to 1,1,0,0,1, and first columns that 1} is encoded into table 2 according to 1,1,0,0,1,1,0,0}.Table 3 is that table 1 is finally modulated-the array code coding result.For example: the element of the 9th row of first row, the h among the corresponding diagram 1C (1), so h (1)=1 0 0 0 0 0 0 0 1 is h (1)=0.

Decode procedure: table 4 is to deliver to the reception data that decoder will be deciphered, and is a modulation-array code code word.Maximum is 255, and minimum value is 0.Table 5 is that modulation code part code element is that unit sorts with 6: 8 equivalent weight code code words, the numeral code element at end descending order in code word, and example first row code element ordering back amplitude size is followed successively by 162,149,135,107,55,0,0,0, so table 5 first is classified { 162-0,135-2 as, 0-5,55-4,107-3,149-1,0-6,0-7}.Can draw threshold value by-3 and-4 elements { (107,55), (173,162), (155,79), (224,52), (255,88), (223,55), (255,77), (169,66) } of every row simultaneously is 134.Table 6 is soft-decision metrics primary Calculation results, for example first be listed as, as shown in Figure 10 the 162-0 of table 5 and 55-4 calculate respectively table 6 162-0 and-(255-55)-4, the nine row, as shown in Figure 11 83 of table 5 and 199 calculate respectively table 6-(255-83) and 199.Table 7 is the revised results of the modulation code part code element soft-decision metrics first step, as first row, as shown in Figure 12 the 162-0 of table 6 and-200-4 be modified to 162-0 and-320-4.Table 8 is revised results of modulation code part code element soft-decision metrics second step, as first row, as shown in Figure 13 the 171-3 of table 7 and-320-4 be modified to 171-3 and-171-4.Table 9 is revised results of modulation code part code element soft-decision metrics the 3rd step, as first row, start=5 and end=7 as shown in Figure 14, Figure 15 as can be known table 8-357-5 ,-306-6 and-255-7 is modified to-255-5 ,-255-6-0 and-255-7.Table 10 is results behind the iterative decoding, and table 11 is bit decision output results.

Table 1:

????1	????0	????0	????0	????0	????0	????0	????0
								????1	????1	????1	????1	????1	????1	????1	????1
????0	????1	????0	????0	????0	????1	????0	????0
								????0	????1	????0	????1	????0	????1	????0	????1
????1	????1	????1	????1	????1	????1	????0	????1
								????1	????0	????0	????0	????1	????1	????0	????0

Table 2:

????1	????0	????0	????0	????0	????0	????0	????0
								????1	????1	????1	????1	????1	????1	????1	????1
????0	????1	????0	????0	????0	????0	????1	????0
								????0	????1	????0	????1	????0	????1	????1	????1
????1	????1	????1	????1	????1	????1	????0	????1
								????1	????0	????0	????0	????1	????0	????0	????0
????0	????0	????1	????0	????0	????1	????1	????0
								????0	????0	????1	????1	????1	????0	????0	????1

Table 3:

????1	????0	????0	????0	????0	????0	????0	????0	????0	????1
										????1	????1	????1	????1	????1	????1	????1	????1	????1	????0
????0	????1	????0	????0	????0	????0	????1	????0	????1	????0
										????0	????1	????0	????1	????0	????1	????1	????1	????0	????0
????1	????1	????1	????1	????1	????1	????0	????1	????0	????1
										????1	????0	????0	????0	????1	????0	????0	????0	????1	????0
????0	????0	????1	????0	????0	????1	????1	????0	????0	????1
										????0	????0	????1	????1	????1	????0	????0	????1	????1	????1

Table 4:

????162	????6	????43	????51	????41	????55	????39	????66	????83	????91
										????135	????196	????217	????255	????255	????255	????255	????255	????199	????96
????0	????255	????27	????21	????77	????41	????255	????0	????255	????52
										????55	????255	????79	????255	????83	????223	????255	????255	????55	????20
????107	????173	????243	????224	????255	????255	????30	????169	????49	????120
										????149	????46	????51	????48	????255	????26	????77	????49	????181	????22
????0	????63	????255	????52	????88	????255	????255	????49	????38	????255
										????0	????162	????155	????255	????255	????0	????0	????191	????242	????123

Table 5:

????162-0	??6-7	??43-6	??51-5	??41-7	??55-4	??39-5	??66-4	????83	????91
										????135-2	??196-2	??217-2	??255-0	??255-0	??255-0	??255-0	??255-0	????199	????96
????0-5	??255-0	??27-7	??21-7	??77-6	??41-5	??255-1	??0-7	????255	????52
										????55-4	??255-1	??79-4	??255-1	??83-5	??223-3	??255-2	??255-1	????55	????20
????107-3	??173-3	??243-1	??224-3	??255-1	??255-1	??30-6	??169-3	????49	????120
										????149-1	??46-6	??51-5	??48-6	??255-2	??26-6	??77-4	??49-5	????181	????22
????0-6	??63-5	??255-0	??52-4	??88-4	??255-2	??255-3	??49-6	????38	????255
										????0-7	??162-4	??155-3	??255-2	??255-3	??0-7	??0-7	??191-2	????242	????123

Table 6:

??162-0	??-249-7	??-212-6	??-204-5	??-214-7	??-200-4	??-216-5	??-189-4	????-172	????-164
										??135-2	??196-2	??217-2	??255-0	??255-0	??255-0	??255-0	??255-0	????199	????-159
??-255-5	??255-0	??-228-7	??-234-7	??-178-6	??-214-5	??255-1	??-255-7	????255	????-203
										??-200-4	??255-1	??-176-4	??255-1	??-172-5	??223-3	??255-2	??255-1	????-200	????-235
??107-3	??173-3	??243-1	??224-3	??255-1	??255-1	??-225-6	??169-3	????-206	????-135
										??149-1	??-209-6	??-204-5	??-207-6	??255-2	??-229-6	??-178-4	??-206-5	????181	????-233
??-255-6	??-192-5	??255-0	??-203-4	??-167-4	??255-2	??255-3	??-206-6	????-217	????255
										??-255-7	??-93-4	??155-3	??255-2	??255-3	??-255-7	??-255-7	??191-2	????242	????-132

Table 7:

??162-0	??-249-7	??-254-6	??-286-5	??-214-7	??-320-4	??-302-5	??-302-4	????-172	????-164
										??189-2	??274-2	??304-2	??255-0	??255-0	??255-0	??255-0	??255-0	????199	????-159
??-357-5	??255-0	??-228-7	??-234-7	??-214-6	??-300-5	??306-1	??-255-7	????255	????-203
										??-320-4	??306-1	??-282-4	??306-1	??-241-5	??357-3	??357-2	??306-1	????-200	????-235
??171-3	??277-3	??292-1	??358-3	??306-1	??306-1	??-270-6	??270-3	????-206	????-135
										??179-1	??-251-6	??-286-5	??-248-6	??357-2	??-275-6	??-285-4	??-288-5	????181	????-233
??-306-6	??-269-5	??255-0	??-325-4	??-267-4	??357-2	??408-3	??-247-6	????-217	????255
										??-255-7	??-149-4	??248-3	??357-2	??408-3	??-255-7	??-255-7	??267-2	????242	????-132

Table 8:

??162-0	??-249-7	??-254-6	??-286-5	??-214-7	??-320-4	??-302-5	??-270-4	????-172	????-164
										??189-2	??274-2	??304-2	??255-0	??255-0	??255-0	??255-0	??255-0	????199	????-159
??-357-5	??255-0	??-228-7	??-234-7	??-214-6	??-300-5	??306-1	??-255-7	????255	????-203
										??-171-4	??306-1	??-248-4	??306-1	??-241-5	??320-3	??357-2	??306-1	????-200	????-235
??171-3	??149-3	??292-1	??325-3	??306-1	??306-1	??-270-6	??270-3	????-206	????-135
										??179-1	??-251-6	??-286-5	??-248-6	??357-2	??-275-6	??-285-4	??-288-5	????181	????-233
??-306-6	??-269-5	??255-0	??-325-4	??-267-4	??357-2	??285-3	??-247-6	????-217	????255
										??-255-7	??-149-4	??248-3	??357-2	??267-3	??-255-7	??-255-7	??267-2	????242	????-132

Table 9:

??162-0	??-249-7	??-254-6	??-286-5	??-214-7	??-320-4	??-302-5	??-270-4	????-172	????-164
										??189-2	??274-2	??304-2	??255-0	??255-0	??255-0	??255-0	??255-0	????199	????-159
??-255-5	??255-0	??-228-7	??-234-7	??-214-6	??-300-5	??255-1	??-255-7	????255	????-203
										??-171-4	??255-1	??-248-4	??255-1	??-241-5	??320-3	??255-2	??255-1	????-200	????-235
??171-3	??149-3	??292-1	??325-3	??255-1	??255-1	??-270-6	??270-3	???-206	????-135
										??179-1	??-251-6	??-286-5	??-248-6	??255-2	??-275-6	??-285-4	??-247-5	????181	????-233
??-255-6	??-269-5	??255-0	??-325-4	??-267-4	??255-2	??255-3	??-247-6	????-217	????255
										??-255-7	??-149-4	??248-3	??255-2	??255-3	??-255-7	??-255-7	??267-2	????242	????-132

Table 10:

????77	????-115	????-503	????-516	????-77	????-502	????-503	????-427	????-172	????-164
										????458	????230	????281	????616	????598	????230	????549	????568	????199	????-159
????-522	????478	????-166	????-166	????-528	????-596	????184	????-503	????255	????-203
										????-446	????549	????-498	????199	????-199	????662	????579	????211	????-200	????-235
????127	????419	????530	????519	????127	????159	????-557	????539	????-206	????-135
										????521	????-225	????-579	????-561	????524	????-225	????-262	????-608	????181	????-233
????-597	????-592	????211	????-599	????-561	????505	????206	????-206	????-217	????255
										????-226	????-503	????584	????223	????542	????-561	????-517	????223	????242	????-132

Table 11:

????1	????0	????0	????0	????0	????0	????0	????0
								????1	????1	????1	????1	????1	????1	????1	????1
????0	????1	????0	????0	????0	????1	????0	????0
								????0	????1	????0	????1	????0	????1	????0	????1
????1	????1	????1	????1	????1	????1	????0	????1
								????1	????0	????0	????0	????1	????1	????0	????0

The experimental result preliminary by Figure 16 can show: as long as the original error rate (being defined as the error rate that adopts after overall optimal threshold quantizes) is lower than 4 * 10 ^-2, the result of use of 8 * 10 modulation-array codes just can with 1: 2 modulation code (the two code check is suitable) result of use suitable; When the original error rate is lower than 1 * 10 ^-2After, its result of use significantly is better than 1: 2 modulation code.So method and device proposed by the invention have reached intended purposes.

Claims (6)

1, the coding method of the modulation-array code used of a kind of volume hologram memory contains modulation coding method and error correction/encoding method, it is characterized in that it is made of in turn following steps:

1) make the storage data carry out the equivalent weight code coding by the equivalent weight code encoder;

2) carry out error correcting code coding as input by the error correcting code encoder with equivalent weight code, the error correcting code of employing is a systematic code;

3) the code element structure behind the coding from left to right is made up of modulation code part and verification unit part successively, and wherein, left side modulation code partly is: k * k code element, from left to right be: x (1,1), x (1,2) ... x (1, k) }, ... { x (k, 1), x (k, 2), ... x (k, k) } be that k n:k modulation code code word arranged in order.The right part: k of verification unit * 2 code elements, h (1) wherein, h (2), ... h (k-1), h (k) with the pass of left side sign indicating number is: (1, i) x (2 for h (i)=x, i) ... x (k-1, i) x (k, i) 1 i=1,2 ... k-1, k, d (1), d (2) ... d (k-1), d (k), with the pass of left side sign indicating number be: (1, i) f (2 for d (i)=f, i) ... f (k-1, i) f (k, i) 1 i=1,2 ... k-1, k, f (j, i)=x (j, ((i+j-2) mod k)+1) j=1,2 ... k-1, k.

2, the coding method of the modulation-array code used of volume hologram memory according to claim 1 is characterized in that: adopt 6: 8 equivalent weight codes in the equivalent weight code described in the encoder of routine.

3, the coding method of the modulation-array code used of volume hologram memory according to claim 1, it is characterized in that: adopted dsp chip to be designed to coding structure encoder of the present invention, described error correcting code adopts the modified model array code of being made up of two (9,8) parity check codes.

4, the coding method of the modulation-array code used of volume hologram memory according to claim 1 and the coded system that designs, it is characterized in that: it contains oriented 6: 8 equivalent weight code encoders provides the data source of storing data, 6: 8 equivalent weight code encoders of coded data are provided to the systematic error checking code encoder, provide the systematic error checking code encoder of coded data, the spatial light modulator of the coded data after storage crystal provides modulation and storage crystal to spatial light modulator.

5, the interpretation method of the modulation-array code used of a kind of volume hologram memory contains the soft-decision metrics algorithm, and it is characterized in that: it is made of successively following steps:

1) data of coming from storage crystal are read in data card;

2) computer is read a code word from data card;

3) enter the decoding subprogram;

4) be the code element of the part of the modulation code in the code word that unit sorts from big to small by the amplitude that receives with the equivalent weight code code word, add up overall discrimination threshold;

5) enter the soft-decision metrics counting subroutine: at first the separated into two parts code element is calculated respectively, wherein modulation code part code element enters modulation code part soft-decision metrics primary Calculation subprogram, verification unit part code element enters verification unit part soft-decision metrics counting subroutine, and the two enters modulation code part code element soft-decision metrics patch routine after calculating and finishing;

6) the modulation code part soft-decision metrics primary Calculation subprogram of being mentioned 5), wherein the primary Calculation formula is:

S(u|y)＝c×|b+d×y|

C representative is 0 or 1 o'clock positive-negative polarity information with bit decision in the formula, and d and b represent the various complex informations of channel,

If: receiving data maximum after putting in order is Max, and minimum value is 0,

Then: boundary condition is:

When judging u=1, if the confidence level minimum that y=0 then judges, | S (u|y) |=0, if the confidence level maximum that y=Max then judges, | S (u|y) |=Max,

When judging u=0, if the confidence level maximum that y=0 then judges, | S (u|y) |=Max,

If the confidence level minimum that y=Max then judges, | S (u|y) |=0,

Draw:

C in the formula ₁, c ₀The expression code element get 1 and 0 o'clock S (u|y) polarity, c wherein ₁* c ₀=-1;

7) the verification unit part soft-decision metrics counting subroutine of 5) being mentioned, wherein threshold value is 4) in the global decision threshold value mentioned, in modulation code part code element sequencer procedure, take out the m1 and (m1-1) range value of individual code element in each equivalent weight code code word, search out the threshold value that makes these code elements differentiate back error code minimum, the best global threshold of the modulation code part code element that Here it is counts, it is approximated to the decision threshold of verification unit part code element, so verification unit part soft-decision metrics computing formula is:

8) the modulation code part code element soft-decision metrics patch routine of being mentioned 5) divided for three steps carried out at least, wherein: at first enter the soft-decision metrics first step patch routine of modulation code part code element, revise according to the position after the ordering in the modulation code word successively:

Wherein:

f(y，n)＝y×g(n)

N is the position after code element sorts in modulation code word, n=0, and 1,2 ..., k-1 then enters second step of modulation code part code element soft-decision metrics patch routine, is revised according to the bit error probability of code element:

S _m1-1(u|y)＝c ₁×min(|S _m1-1(u|y)|，|S _m1(u|y)|)

S _m1(u|y)＝c ₀×|S _m1-1(u|y)|

M1 represents in the modulation code code word 1 number, S in the formula _n(u|y) represent the soft-decision metrics of n code element, n is the position after code element sorts in modulation code word, n=0,1,2 ..., k-1, carry out the 3rd step of the soft-decision metrics patch routine of modulation code part code element then, revise according to the reception range value of code element: as n, n+1, ..., the y value of n+m code element is identical, and this m+1 code element polarity is identical, then:

S _n(u|y)＝...＝S _n+m(u|y)＝sgn(S _n(u|y))×min(|S _n(u|y)|，...，|S _n+m(u|y)|)

If: the polarity difference of m+1 code element, then:

S _n(u|y)＝S _n+1(u|y)＝...＝S _n+m(u|y)＝0

Thereby obtain the condition soft-decision metrics value of modulation code part code element;

9) carry out the iterative decoding of code word, wherein the operation rule of soft-decision metrics is as follows:

S(u1u2)≈sgn(S(u1|y1))×sgn(S(u2|y2))×min(|S(u1|y1)|，|S(u2|y2)|)

Sgn () gets sign function in the formula, and min () is the function of minimizing;

10) the final soft-decision metrics according to code element quantizes, and its polarity is c ₁Code element be 1, polarity is c ₀Code element be 0, and then it is carried out equivalent weight code decoding, the court verdict data after the output decoding at last;

11) judgement finishes to obtain storing data after data card will be adjudicated output.

6, the decoding system that designs of the interpretation method of the modulation-array code of using according to the volume hologram memory of claim 5, it is characterized in that: it comprises storage crystal, scan the charge coupled device ccd that receives data from storage crystal, filter filtering is carried out in output to CCD, the computer that links to each other with filter output successively, computer have and receive and the data card of dateout.

Images